Could you give us some more details please. ?
My M-DAC:
The oscilloscope showed the instability of the digital DC (U5 /1V2DC, 3V3D) ESS9018, which depended on the presence of the input signal and the selected input. Replacement of all Lelon OCRZ series capacitors in digital DC is not effective.
IMXO, for perfectionist reasons, you can replace the capacitors of the OCRZ series. In addition to digital, they shunt the audio part of the ESS9018
Capacitors of this type are very effective on the HF range, but are much less effective on the LF (sound) range.
Datasheet. Ripple current (RC)
Lelon OCRZ series - RC(120Hz) = 0.05 RC (100kHz). And RC(20Hz) = 0.0?? RC(100kHz).
FPCAP Nichicon capacitors are 2 times more efficient in this parameter. RC (120Hz) = 0.1 RC (100kHz).
For example: Panasonic FC series, RC (60Hz) = 0,55-0,8 RC (100 kHz).
Tweak's internal DC digital stabilizers solved this problem.
This is just my experience. It is not a recommendation for repetition. Not responsible for any consequences.
All signs from the M-DAC service manual.
- Replaced the U5 NE5532D with AD8066, the most suitable one that was at hand.
- Changed the output voltage monitoring points.
- Radically reduced the values of resistors R319, R321 (the values may depend on the selected chip).
Oscilloscope stability chip control is mandatory.
- I installed additional shunt capacitors directly on the chips of the stabilizers.
External power 5 rails clean DC can not solve this problem.
Best Regards